- Email: [email protected]
Pathovar Discrimination within Pseudomonas syringae subsp. savastanoi Using Whole Cell Fatty Acids and Pathogenicity as Criteria
System. Appl. Microbiol. 14, 79-84 (1991) © Gustav Fischer Verlag, StuttgartlNew York
Pathovar Discrimination within Pseudomonas syringae subsp. savastanoi Using Whole Cell Fatty Acids and Pathogenicity as Criteria J. D. JANSE Department of Bacteriology, Plant Protection Service, 6700 He Wageningen, the Netherlands Received December 14, 1989
Summary Fifry-two strains of the bacterium Pseudomonas syringae subsp. savastanoi from six different hosts were characterized by physiological, biochemical and pathogenicity tests and whole cell fatty acid analysis (FAA). Six other Pseudomonas strains were included for comparison in FAA. All 52 strains formed a very homogenous group on the basis of physiological and biochemical tests. On the basis of pathogenicity tests on 4 different hosts, strains could be divided into the three pathovars of P. syringae subsp. savastanoi formerly described, viz. pv. oleae, pv. nerii and pv. fraxini. This division into pathovars could be confirmed by FAA, thus showing that FAA is a supportive tool for taxonomic study below species level of plant pathogenic bacteria. Pv. fraxini strains (isolated from Fraxinus excelsior), formed a discrete cluster in principal component analysis. Pv. nerii strains (isolated from Nerium oleander) and pv. oleae strains (isolated from Olea europea, Jasminum sp., Ligustrum japonicum or Phillyrea sp.) could also be separated but showed close relationship. Two pv. fraxini strains showed a deviating (pv. oleae) fatty acid pattern. The fatty acids 11 : 0 iso 30H, 20: 0 and 20: 1 trans 11 were found to be exclusively present in pv. nerii, but they were not present in all strains. Pathovar discriminative fatty acids were 16: 0, 16: 1 cis 9, 18: 0, 18: 1 cis 11 or the ratio of 16: 1 cis 9118: 1 or 16: 1 cis 9 + 16:0118: 1 + 12:0 20H.
Key words: GLC - Multivariate analysis
Introduction Pseudomonas syringae subsp. savastanoi (Smith) Janse is a plant pathogenic bacterium, causing necrotic excrescences or galls on species of the Oleaceae and on Nerium oleander L. (Apocynaceae). On the basis of host range, pathogenesis and plant hormone production, three pathogenic varieties (pathovars) of this bacterium have been described Uanse 1981 a, b, 1982 a, b): pv. oleae, causing parenchymatous galls on various species of the Oleaceae, pv. nerii, evoking parenchymatous galls or wartlike excrescences on N. oleander and various species of the Oleaceae and pv. fraxini, causing necrotic, wartlike excrescences on Fraxinus excelsior L. and to some extent on Olea europea L. The first two pathovars produce plant hormones, viz. cytokinins and indole acetic acid, lAA (Evidente et aI., 1986; Smidt and Kosuge, 1978). The information for lAA production of pv. nerii appears to be located on a plasmid, that of pv. oleae on the chromosome
(Comai et aI., 1982). Pv. fraxini does not produce plant hormones Uanse, 1981 a). The discrimination between the pathovars of subsp. savastanoi, based on the above mentioned characters, is complicated and/or time consuming. A preliminary study by Varvaro and Sasser (1987) indicated that pv. oleae and pv. nerii could be separated by whole cell fatty acid profiles. They did not study strains of pv. fraxini, however. Wells et ai. (1990), also using whole cell fatty acid analysis, could distinguish pv. fraxini from the other pathovars but notpv. oleae from pv. nerii. Fatty acid analysis by gasliquid chromatography, especially after the development of automated systems (Miller and Berger, 1985) is a rapid and stable method for taxonomic and identification work. Therefore I tried 1) to differentiate a large number of strains of subsp. savastanoi into fatty acid groups, eventually correlating with known pathovars, 2) to establish a
J. D. Janse
80
library for future reference and identification. Some (serologically) related Pseudomonas strains were included for comparison. In this study the Classification of 52 strains of subsp. savastanoi from six different hosts, using phenotypic biochemical and pathogenicity tests and fatty acid' analysis is reported. Furthermore the succesful separation of strains in the three known pathovars on the basis of fatty acid patterns, is demonstrated.
Material and Methods Bacterial strains. Table 1 lists the 58 strains used with their identity and origin. All strains were routinely maintained on
Strain
P. syringae subsp. savastanoi PD l 109, 116 (= PDDCC 2 7711), PD119, 120, 159 (= PDDCC7712), PD160, 161,179, 180,206, PD316, 392,547,548, 555 NCPPB 3 1006 (= PD124) NCPPB1464 (= PD122) CFBp4 1838 (= PD166) L39-4 (= PD532)
nutrient agar (NA, Difco) with 0.1 % w/v D-( + )-glucose, and all strains were lyophilized. Physiological and biochemical properties. Methods used by Janse (1981 b) were applied. Pathogenicity tests. Two-year old seedlings of Fraxinus excelsior L., and rooted cuttings of Olea europea L., Nerium oleander L. and Forsythia intermediiJ Zab. were used. For each isolate, 2-4 plants were used and 5 inoculations carried out per plant, inoculating both vigorously growing sprouts and suberized stems during April/May (F. excelsior L.) or June/July (other plants). A hypodermic needle and a c. 107 cells. ml- 1 suspension in sterile phosphate buffered saline (pH 7.2) of a 24 h NA culture was used for inoculation. After inoculation, F. excelsior and F. intermedia were kept a few days under high humidity (c. 20°C, 95% RH) in a glasshouse whereafter they were grown outdoors. O. europea
Host
Country
Fraxinus excelsior L.
Netherlands
Fraxinus Fraxinus Fraxinus Fraxinus
excelsior excelsior excelsior excelsior
L. L. L. L.
U.K.
UK. France France (L. Gardan)
PD181 (= NCPPB3278), 317, 390, 911,1235 PD1188, 1189, 1299, 1300 L7/12 (= PD529), L14/5 (= PD533) NCPPB640 (= PD125 )
Nerium oleander L.
Spain
Nerium oleander L. Nerium oleander L. Nerium oleander L.
Jordan France (L. Gardan) Yugoslavia
PDIIS7, 1190, 1296, 1297, 1298
Olea europea L.
Jordan
PD912, 913, 914 PDI056 PD1265, 1266 NCPPB639 (= PD118) NCPPB1481 (= PD 186) NCPPB2327 (= P0l21) T38/1 (= PD534), K23 /15 (= PD535)
Olea Olea Olea dlea Olea Olea Olea
CFBP1751 (= PD167) T12/4 = PD531 Phyll (= PD530), Phyl2 (= PD536) NCPPB2328 (= PD185)
Jasminum sp. Jasminum sp. Phyllirea sp. Ligustrum japonicum Thbg
Greece France (L. Gardan) France (L. Gardan) Italy
Persea americana Mill. Syringa vulgaris L.
Israel
P. syringae pv. syringae NCPPB191 (= PD321) NCPPB281 (= PD184)
europea europea europea europea europea europea europea
L. L. L. L. L. L. L.
P. syringae pv. morsprunorum NCPPB560 (= PD213)
Greece Maroc Italy (M. Scortichini) Yugoslavia Yugoslavia Italy France (L. Gardan)
U.K.
U.K.
P. syringae pv. maculicola PD236
Brassica oleracea L.
Netherlands
Pseudomonas species PD117 NCPPB1465 (= PD123)
Fraxinus excelsior L. Fraxinus excelsior L.
Netherlands
1 2
3
4
U.K.
PD, culture collection Plant Protection Service, Wageningen, the Netherlands. PDDCC, Plant Disease Division Culture Collection, Auckland, New Zealand. NCPPB, National C;;llection of Plant Pathogenic Bacteria, Harpenden, U.K. CFBP, Collection Fran~aise de Bacteries Phytopathogenes, Angers, France.
Table 1. Origin and identity of strains used in this study
Pathovars of Pseudomonas syringae and N . oleander were grown in a computer controlled glasshouse at 28°C, 85-90% RH and 10 000 Lux light after inoculation. Fatty acid analysis. Bacteria were grown for 48 h at 28°C on Trypticase Soy Broth Agar (TSBA), containing (w/v) 3% Trypticase Soy Broth (BBL) and 1.5% Bacto Agar (Difco). Circa 40 mg (wet weight) ccells were harvested from the most dilute quadrant showing confluent growth (late log phase). Whole cell fatty acids were liberated, methylated and extracted, following the method of Miller and Berger (1985). All strains were tested in duplicate. The Midi Microbial Identification System MIS (Microbial ID, Inc. Newark, DE, USA) was used. The MIS consists of a Hewlett Packard HP5890A gas chromatograph with a 25 m x 0.2 mm 5% methylphenyl silicone fused silica capillary column, H2 as carrier gas and a flame-ionization detector, an automatic sampler, an integrator and a computer. The latter identifies the fatty acids, using data of a fatty acid library and a calibration mix of known fatty acids (Microbial ID, Inc.). Statistical analysis (standard error of differences of the means) of values for some fatty acids or their ratios was performed using the program GENSTAT on a MicroVAX 3600 computer. Moreover library generating software (LGS) and a statistical program CLUS developed by Microbial ID Inc., was used for principal component and cluster analysis of strains and also for creating a reference library for P. syringae subsp. savastanoi.
Results
Physiological and biochemical tests All strains (PD 117, 123, 184, 213, 236, 321 not included, for their deviatiQns from the pattern described below see janse, 1982 a) behaved similarly in physiological and biochemical tests. They were Gram-negative, motile, non-spore forming rods, showing slow growing 'gray-white, smooth, glistening raised and circular or slightly irregular to undulate colonies on NA. Usually levan-negative on nutrient-sucrose (5% w/v) agar (PD ' 120,161,179,180,206,555 positive). Strains produced a ' weak, blue-green fluorescent, diffusable pigment on King's medium B (CFBP 1751, PD 179 and PD 534 produced a brown diffusable pigment). Glucose metabolism respiratory, oxidase negative, catalase negative. Acid was produced from D-( + )-galactose, glucose, D-( +}-ribose, sucrose (slow), D-( + }-xylose and mannitol. No acid was produced from maltose, D-( + }-raffinose (PD 161 positive), erythritol or salicin. Alkali was produced from L-( + }-tartrate (PD 120, 392, T 12/4 and T 38/1 negative). Aesculin and casein were not or only weakly hydrolyzed, arginine, gelatin, starch and arbutin were not hydrolyzed. Nitrates were not reduced, nor was H2S produced from cysteine. No growth at 37 DC or with 5% NaCl. Sodiumpolypectate was usually hydrolized at pH 5.5 (NCPPB 639, 640,1464, 2327,2328, 3278 and PD 1056 negative). Hypersensitivity was produced in tobacco (cv. Samson) leaves (NCPPB 640, PD 1265, 1266, 1296, 1300 negative, L 7/12, L 24/5, NCPPB 1481, T 12/4 doubtful). On the basis of the above mentioned tests all 52 strains were identical to P. syringae subsp. savastanoi.
Pathogenicity tests Results of pathogenicity tests on four different hosts are presented in Table 2. At least three distinct groups could 6 System. AppJ. Microbial. Vol. 14/1
81
Table 2. Results of pathogenicity tests with 52 sttains of P. syringae subsp. savastanoi on 4 different hosts Strains from F. excelsior O.europea N. 0leander3 Jasminum Ligustrum Phyllirea
F. excelsior
+
Et>1 Et>2 Et> Et>
O. europea
Pathogenicity on: N. F. interoleander media
+
Et> Et> Et>4 Et> Et>
Et>1+
+, necrotic swellings; Et>, parenchymatous galls; -, no pathoreaction Strain PD1266 and 1298 doubtful; PD1296 negative. Strain PD1300 negative. Strain NCPPB640 non pathogenic. Strain T12/4 doubtful.
~enic 2
3 4
be observed, viz. 1} strains from Oleaceae other than F. excelsior, causing parenchymatous galls on O. europea and F. excelsior (the strains from jasminum and Ligustrum, also causing parenchymatous galls on F. intermedia included), 2} strains from N. oleander causing parenchymatous galls or wartlike excrescences on N. oleander, O. europea and F. excelsior and 3} strains from F. excelsior causing necrotic excrescences on F. excelsior and O . europea. On the basis of these results strains from O. europea, jasminum, Ligustrum japonicum and Phillyrea could be classified as pv. oleae, all strains from F. excelsior as pv. fraxini and all strains from N. oleander as pv. nerii, confirming earlier results.
Fatty acid analysis Table 3 shows the mean percentages of fatty acids found in the tested strains of P. syringae subsp. savastanoi. Over 83 % consisted of the straight chain saturated and monounsaturated fatty acids 12:0 (c. 4.5%), 16:0 (c. 27%), 16:1 cis 9 (c. 31%) 18:1 cis 11 (c. 20%) and 18:0 (c. 1.5%). About 10.5% were the hydroxy fatty acids 10:0 30H, 12: 0 20H and 12: 0 30H. The cyclic fatty acid 17: 0 was found to be present in rather large amounts (c. 4%) . The above mentioned fatty acids were found in all strains. The other 14 fatty acids (in total c. 1.5%) occurred infrequently. The fatty acids 15: 0 (1 X NCPPB 2327}, 17: 0 and 19: 1 trans 7 have never been reported before to occur in P. syringae subsp. savastanoi. The fatty acids 11 : 0 iso 30H, 19: 0 iso, 20: 1 trans 11 and 20: 0 occurred in strains from N. oleander only, they were not present in all Nerium strains, however. The fatty acid pattern of the other Pseudomonas strains differed to a large extent from that of P. syringae subsp. savastanoi (Table 4) resulting in low similartiy values (below 0.5) with the P. s. subsp. savastanoi library. Statistical significant differences between concentrations of fatty acids or their ratios (Table 5) were correlated with the segregation of strains in three groups (pathovars) which was found in the pathogenicity tests.
82
J.D.Janse
Fatty acid saturated 12:0' 14:0' 15:0' 16:0' 17:0' 18:0' 20':0' branched 17:0' iso 19:0' iso cyclo 17:0' cy 19:O'cy Cll-12
count
mean%
s.d.
10'4 63
4.5 0'.2 t 27.1 t 1.5
0'04 0'.2
1
10'4 17 10'3 2 44 1
10'3 8
0'.3
3.3 0'.8
0'04
Table 3. Total cellular fatty acid profile of 52 strains* of Pseudomonas syringae subsp. savastanoi grown for 48 h on TSBA medium
s.d.
count
hydroxy 10':0' 30H 11:0' iso 30H 12:0' 20H 12:0' 30H
10'4 8 10'4 10'4
3.1 t 3.2 4.5
0'.2 0'.3
unsaturated 16:1 cis 9 18: 1 cis 11 19:1 trans 7 20':1 trans 11
10'4 10'4 5 7
31.6 19.8
3.3 5.8
56
0'.4
0'04
unknown! 19:0' cydo C 9-10' 3.7
mean%
fatty acid
0'.3
2.0'5
.t
- = not detected; t = trace
* All strains tested in duplicate.
Table 4. Percentage of several fatty acids of some Pseudomonas bacteria* related to P. syringae subsp. savastanoi Fatty acid
Bacterium
P. syringae pv. syringae (NCPPB191) P. syringae pv. syringae (NCPPB281) P. syringae pv. morsprunorum (NCPPB56D) P. syringae pv. maculicola (PD236) Pseudomonas sp. (PD117) Pseudomonas sp. (NCPPB1465) P. syringae subsp. savastanoi
12:0' 20H
16:0'
16:1 cis 9
18:0'
18:1 cis 11
3.7 3.7 3.5 3.0' 0' 0' 3.2
20'.7 25.0' 2604 25.5 25.7 19.6 27.1
36.1 34.7 30'.6 36.9 26.0' 4.9 31.6
0'.5 1.2 1.5
17.0' 17.6 17.4 18.4 18.7 32.7 19.8
1.2
0'.2 0'.9 1.5
.. Result of a test in duplicate.
Fatty acid
pv. fraxini mean %
pv. nerii mean %
pv. oleae mean %
s.e.d.
16:0' 16:1 18:0' 18:1 16:1 16:1
31.0'3 34.81 0'.98 12.69 2.82 4.24
23.85 28.56 2.31 25.86 1.16 1.88
25.47 30'.28 1.64 22.15 1.37 2.21
0'0471 0'.634 0'.162 0'.755 0'.0'89 0'.118
1
cis 9 cis 11 cis 9/18:1 cis 9+16:0'118:1+12:0' 20H
Table 5. Fatty acids or their ratios that differentiate pathovars of P. syringae subsp. savastanoi (P = 0'.0'5)1
Based on data of 50' strains tested in duplicate (deviating strains NCPPB 10'0'6 and CFBP 1838 excluded).
There was greater difference between strains from F. excelsior and those from other hosts than between strains from N. oleander and those from Oleaceae other than F. excelsior. Only four deviating strains were found, two strains from F. excelsior (NCPPB 1006 and CFBP 1848) repeatedly showed i fatty acid pattern of pv. oleae, one strain from N. oleander (PD 1300) showed a pattern of pv.
oleae and one strain from olive (NCPPB 1481) showed a pattern of pv. nerii. Principal component analysis showed that pv. fraxini formed a separate group, but that pv. oleae and nerii had an overlap for these features (Fig. 1). This suggests a closer relationship between pv. oleae and pv. nerii than between pv. fraxini and the other two pathovars.
Pathovars of Pseudomonas syringae
....c: -11.2 N
OJ
c: o
a. E
8 -13.1 ~
a.
o
I.J
c:
'.£ -15.0 -16.9
-16.6
-20.8
-24.6
•••
• • ••• •• • •
...
~----~--~~--~~--~----~
-21.3
-14.6
-7.9
-1.3
-5.4
-12.1
Principle component 1 E~clidian
distance
Fig. 1. Two-dimensional plot of principle component analysis of 52 strains of Pseudomonas syringae subsp. savastanoi (tested in duplo) showing subdivision in three fatty acid groups. These groups are corresponding to pv. fraxini (e), pv. oleae (0) and pv. . nerii (+).
Discussion This study shows once more that fatty acid analysis can be a supportive tool for the classification of plant pathogenic bacteria below species level, as was also reported by De Boer and Sasser (1986) for subspecies of Erwinia carotovora. Strains of P. syringae subsp. savastanoi allocated to pv. oleae or pv. nerii based on pathogenicity and host range could be separated into two similar groups by fatty acid analysis, confirming results of Varvaro and Sasser (1987). Wells et aL (1990) could not separate pv. oleae from pv. nerii, possibly because they used a different medium, six-day old cells, a different column and a different chromatograph. The differences in fatty acid profile of pv. oleae and pv. nerii correlated well with those found by Varvaro and Sasser (1986), even though they used 24 h old cells instead of 48 h old cells. I found, however, that many strains do not produce reproducable amounts of usable cell material after 24 h, yielding variable profiles. The acids 11: 0 iso 30H, 20: 0 and 20: 1 trans 11 were also found by me to be exclusively present in {?v. nerii. However 17: 0 iso was also found to occur after 48 h cultivation in pv. oleae and to a minor extent in pv. fraxini. While strains from L.
83
;aponicum and Jasminum showed some differences in host range as compared to strains from Olea europea, these strains all produced parenchymatous galls and showed a fatty acid profile similar to that of olive strains. Varvaro and Sasser (1987) found a similar pattern studying a larger group of strains from L. ;aponicum. Therefore no further subdivision into pathovars was thought to be useful and justified at the moment. The occurrence of new pathovars may not be excluded however since a new host has cecently been described, viz. Myrtus sp.( Gardan and Abu Ghprrah, 1987). The strains allocated to pv. fraxini on the basis of pathogenicity and host range could be clearly separated from those belonging to pv. nerii or pv. oleae, confirming results of Wells et aL (1990). Several deviating strains were found, not fitting in the pathovar pattern, as far as their fatty acids were concerned. Two strains from F. excelsior, one from France and one from the U.K. showed a pv. oleae profile. Other F. excelsior strains from these countries showed the pv. fraxini pattern. The basis and frequence of this deviation are unknown and need further study, other characters were consistent for pv. fraxini. Strains with the pathogenic phenotype of pv. nerii have been isolated from olive (Pyrolowakis and Welzien, 1974; Wilson and Magie, 1963). The strain isolated from pv. nerii phenotype was from this area, however, it did not show the pathogenic phenotype of pv. nerii. This means that deviating strains (i. e. host not correlating with pathovar classification in fatty analysis), can only be definitely identified by a pathogenicity test on different hosts. This study on fatty acids thus confirms that the existence of the three pathovars previously described Uanse, 1982 b) can be confirmed by fatty acid analysis. It must be remarked that pv. nerii and pv. oleae upon principal component analysis showed a close relationship and a partly overlap. Pv. fraxini was much more different. This correlates with the pathogenic properties and geographic distribution of the different pathovars. Pvs nerii and oleae both produce parenchymatous galls and growth promoting hormones (Surico et aL, 1985). They occur mainly in the Middle East/Southern Europe area. Pv. fraxini produces necrotic excrescences and no hormones, it occurs mainly in the MiddlelW'estern Europe area. This study also shows that the pathovars of P. syringae subsp. fraxini show more differential characters and are therefore more natural (polythetic) than originally suspected. The definition of pathovar used here is that of the Bacteriological Code (Lapage et aI., 1975): pathogenic to one or more hosts, and not the one used by Dye et al., (1980) where it has almost the definition of subspecies. When using the former definition, there would be no need for designating type strains of pathovars (pathotypes, as defined by Dye et al. 1980) since the concept of pathovars does not fall under the rules of the Code. Pathotypes, however, were mentioned by me in an abstract Uanse, 1987) and because the pathovars could be eventually elevated in rank, they are repeated here for valid publication: strain PD 118 = NCPPB 639 = ATCC 13522 as pathotype for pv. oleae
84
J.D.Janse
(also type strain for subspecies savastanoi) strain PD 181 = NCPPB 3278 as lectopathotype for pv. nerii strain PD 116 = PDDCC 7711 as pathotype for pv. fraxini When several (phenotypically or serologically) related Pseudomonas strains were compared in fatty acid analysis, P. syringae subsp. savastanoi could be clearly differentiated from them. This was also true for a saprophytic strain (NCPPB 1465) which was originally described as a deviating pathogenic strain, namely P. savastanoi subsp. fraxini (Sutic and Dowson, 1963). Moreover a hypersensitivity positive P. syringae strain isolated from a lesion on F. excelsior with a close though deviating phenotypic profile, serologically cross-reacting, but not pathogenic for ash could be easily differentiated by fatty acid analysis from subsp. savastanoi Uanse, unpubl.). These findings suggest caution when postulating all kind of transitions in the P. syringae group especially when studying strains from the phyllosphere and not from active lesions or galls (Ercolani, 1983). Acknowledgements: I wish to extend my thanks to J. H. J. Derks, B. E. Spit and W. H. van der Tuin for skilful technical assistance and Dr. j. Dekker and Dr. M. Sasser for critical reading of the manuscript.
References Comai, L., Surico, G., Kosuge, T.: Relation of plasmid DNA to indoleacetic acid production in different strains of Pseudomonas syringae pv. savastanoi. J. Gen. Microbiol. 128, 2157-2163 (1982) De Boer, S. H., Sasser, M.: Differentiation of Erwinia carotovora spp. carotovora and E. carotovora spp. atroseptica on the basis of cellular fatty acid composition. Can. J. Microbiol. 32, 796-800 (1986) , Dye, D. W., Bradbury, j. F., Goto, M., Hayward, A. c., Lelliot, R. A., Schroth, M. N.: International standards for naming pathovars of phytopathogenic bacteria and a list of pathovars names and pathotype strains. Rev. Plant Path. 59, 153-168 (1980) Ercolani, G. L.: Variability among isolates of Pseudomonas syringae pv. savastanoi from the phyllosphere of the olive. J. Gen. Microbiol. 129, 901-916 (1983) Evidente, A., Surico, G., Iacobellis, N. S., Randazzo, G.: 1'methyl zeatin, an additional cytokinin from Pseudomonas syringae pv. savastanoi. Phytochemistry 25, 525-526 (1986) Gardan, L., Abu Ghorra, M.: Identification and pathogenicity of Pseudomonas syringae pv. savastanoi. Proceedings 3rd International Working Group on Pseudomonas syringae pathovars.
Ir.
J.
Ministro da Agricultura, Instituto Nacional de Investigacao Cientifica, LisbonIPortugal (1987) janse, J. D.: The bacterial disease of ash (Fraxinus excelsior), caused by Pseudomonas syringae subsp. savastanoi pv. fraxini. I. History, occurrence and symptoms. Europ. J. Forest Path. 11,306-315 (1981a) janse, J. D.: The bacterial disease of ash (Fraxinus excelsior), caused by Pseudomonas syringae subsp. savastanoi pv. fraxini. II. Etiology and taxonomic considerations. Europ. J. Forest Path. 11,425-438 (1981 b). , janse, j. D.: The bacterial disease of ash (Fraxinus excelsior), caused by Pseudomonas syringae subsp. savastanoi pv. fraxini. III. Pathogenesis. Europ. J. Forest Path. 12, 218-231 (1982 a) janse, j. D.: Pseudomonas syringae subsp. savastanoi (ex Smith) subsp. nov., nom. rev., the bacterium causing excrescences on Oleaceae and Nerium oleander L. Int. J. System. Bact. 32, 166-169 (1982 b) janse, j. D.: Lesser known infections of several forms of Pseudomonas syringae. Proceedings 3rd International Working Group on Pseudomonas syringae pathovars. Ministra de Agricultura, Instituto Nacional de Investigacao Cientifica, LisbonIPortugal (1987) Lapage, S. P., Sneath, P. H. A., Lessel, E. F., Skerman, V. B. D., Seeliger, H. P. R., Clark, W. A.: International code of nomenclature of bacteria. Bacteriological Code (1975 revision), Washington DC., American Society of Microbiology 1975 Miller, L., Berger, T.: Bacterial identification by gas chromatography of whole cell fatty acids. Hewlett-Packard Application Note 228-41, 8 pp. (1985) Pyrowolakis, E., Welzien, H. c.: Studies on the distribution of olive-knot, induced by Pseudomonas savastanoi (Sm.) Stev. in the Greek island of Crete. Phytopath. Mediterr. 13, 118-120 (1974) Smidt, M., Kosuge, T.: The role of indole-3-acetic acid accumulation by alpha methyl tryptophan-resistant mutants of Pseudomonas savastanoi in gall formation on oleanders. Physiol. Plant Path. 13,203-214 (1978) Surico, G., Iacobellis, N. S., Sisto, A.: Studies on the role of indole-3-acetic acid and cytokinins in the formation of knots on olive and oleander plants by Pseudomonas syringae pv. savastanoi. Physiol. Plant Path. 26, 309-320 (1985) Sutic, D., Dowson, W. J.: Microbiological characteristics of some isolates and varieties of Pseudomonas savastanoi (Smith) Stevens. Phytopath. Zschr. 49, 156-160 (1963) Varvaro, L., Sasser, M.: Fatty acid profiles of Pseudomonas syringae pv. savastanoi. Proceedings 3rd International Working Group on Pseudomonas syringae pathovars. Ministro de Agricultura, Instituto Nacional de Investigacao Cientifica, LisbonIPortugal (1987) Wells, J. M., Casano, F. J., Surico, G.: Fatty Acid composition of Pseudomonas syringae pv. savastanoi. J. Phytopath. (1990), in press Wilson, E. E., Magie, A. R.: Physiological, serological and pathological evidence that Pseudomonas tonelliana is identical with Pseudomonas savastanoi. Phytopathology 53, 653-659 (1963)
D. janse, Department of Bacteriology, Plant Protection Service, P.O. Box 9201, 6700 HC Wageningen, the Netherlands
Pathovar Discrimination within Pseudomonas syringae subsp. savastanoi Using Whole Cell Fatty Acids and Pathogenicity as Criteria J. D. JANSE Department of Bacteriology, Plant Protection Service, 6700 He Wageningen, the Netherlands Received December 14, 1989
Summary Fifry-two strains of the bacterium Pseudomonas syringae subsp. savastanoi from six different hosts were characterized by physiological, biochemical and pathogenicity tests and whole cell fatty acid analysis (FAA). Six other Pseudomonas strains were included for comparison in FAA. All 52 strains formed a very homogenous group on the basis of physiological and biochemical tests. On the basis of pathogenicity tests on 4 different hosts, strains could be divided into the three pathovars of P. syringae subsp. savastanoi formerly described, viz. pv. oleae, pv. nerii and pv. fraxini. This division into pathovars could be confirmed by FAA, thus showing that FAA is a supportive tool for taxonomic study below species level of plant pathogenic bacteria. Pv. fraxini strains (isolated from Fraxinus excelsior), formed a discrete cluster in principal component analysis. Pv. nerii strains (isolated from Nerium oleander) and pv. oleae strains (isolated from Olea europea, Jasminum sp., Ligustrum japonicum or Phillyrea sp.) could also be separated but showed close relationship. Two pv. fraxini strains showed a deviating (pv. oleae) fatty acid pattern. The fatty acids 11 : 0 iso 30H, 20: 0 and 20: 1 trans 11 were found to be exclusively present in pv. nerii, but they were not present in all strains. Pathovar discriminative fatty acids were 16: 0, 16: 1 cis 9, 18: 0, 18: 1 cis 11 or the ratio of 16: 1 cis 9118: 1 or 16: 1 cis 9 + 16:0118: 1 + 12:0 20H.
Key words: GLC - Multivariate analysis
Introduction Pseudomonas syringae subsp. savastanoi (Smith) Janse is a plant pathogenic bacterium, causing necrotic excrescences or galls on species of the Oleaceae and on Nerium oleander L. (Apocynaceae). On the basis of host range, pathogenesis and plant hormone production, three pathogenic varieties (pathovars) of this bacterium have been described Uanse 1981 a, b, 1982 a, b): pv. oleae, causing parenchymatous galls on various species of the Oleaceae, pv. nerii, evoking parenchymatous galls or wartlike excrescences on N. oleander and various species of the Oleaceae and pv. fraxini, causing necrotic, wartlike excrescences on Fraxinus excelsior L. and to some extent on Olea europea L. The first two pathovars produce plant hormones, viz. cytokinins and indole acetic acid, lAA (Evidente et aI., 1986; Smidt and Kosuge, 1978). The information for lAA production of pv. nerii appears to be located on a plasmid, that of pv. oleae on the chromosome
(Comai et aI., 1982). Pv. fraxini does not produce plant hormones Uanse, 1981 a). The discrimination between the pathovars of subsp. savastanoi, based on the above mentioned characters, is complicated and/or time consuming. A preliminary study by Varvaro and Sasser (1987) indicated that pv. oleae and pv. nerii could be separated by whole cell fatty acid profiles. They did not study strains of pv. fraxini, however. Wells et ai. (1990), also using whole cell fatty acid analysis, could distinguish pv. fraxini from the other pathovars but notpv. oleae from pv. nerii. Fatty acid analysis by gasliquid chromatography, especially after the development of automated systems (Miller and Berger, 1985) is a rapid and stable method for taxonomic and identification work. Therefore I tried 1) to differentiate a large number of strains of subsp. savastanoi into fatty acid groups, eventually correlating with known pathovars, 2) to establish a
J. D. Janse
80
library for future reference and identification. Some (serologically) related Pseudomonas strains were included for comparison. In this study the Classification of 52 strains of subsp. savastanoi from six different hosts, using phenotypic biochemical and pathogenicity tests and fatty acid' analysis is reported. Furthermore the succesful separation of strains in the three known pathovars on the basis of fatty acid patterns, is demonstrated.
Material and Methods Bacterial strains. Table 1 lists the 58 strains used with their identity and origin. All strains were routinely maintained on
Strain
P. syringae subsp. savastanoi PD l 109, 116 (= PDDCC 2 7711), PD119, 120, 159 (= PDDCC7712), PD160, 161,179, 180,206, PD316, 392,547,548, 555 NCPPB 3 1006 (= PD124) NCPPB1464 (= PD122) CFBp4 1838 (= PD166) L39-4 (= PD532)
nutrient agar (NA, Difco) with 0.1 % w/v D-( + )-glucose, and all strains were lyophilized. Physiological and biochemical properties. Methods used by Janse (1981 b) were applied. Pathogenicity tests. Two-year old seedlings of Fraxinus excelsior L., and rooted cuttings of Olea europea L., Nerium oleander L. and Forsythia intermediiJ Zab. were used. For each isolate, 2-4 plants were used and 5 inoculations carried out per plant, inoculating both vigorously growing sprouts and suberized stems during April/May (F. excelsior L.) or June/July (other plants). A hypodermic needle and a c. 107 cells. ml- 1 suspension in sterile phosphate buffered saline (pH 7.2) of a 24 h NA culture was used for inoculation. After inoculation, F. excelsior and F. intermedia were kept a few days under high humidity (c. 20°C, 95% RH) in a glasshouse whereafter they were grown outdoors. O. europea
Host
Country
Fraxinus excelsior L.
Netherlands
Fraxinus Fraxinus Fraxinus Fraxinus
excelsior excelsior excelsior excelsior
L. L. L. L.
U.K.
UK. France France (L. Gardan)
PD181 (= NCPPB3278), 317, 390, 911,1235 PD1188, 1189, 1299, 1300 L7/12 (= PD529), L14/5 (= PD533) NCPPB640 (= PD125 )
Nerium oleander L.
Spain
Nerium oleander L. Nerium oleander L. Nerium oleander L.
Jordan France (L. Gardan) Yugoslavia
PDIIS7, 1190, 1296, 1297, 1298
Olea europea L.
Jordan
PD912, 913, 914 PDI056 PD1265, 1266 NCPPB639 (= PD118) NCPPB1481 (= PD 186) NCPPB2327 (= P0l21) T38/1 (= PD534), K23 /15 (= PD535)
Olea Olea Olea dlea Olea Olea Olea
CFBP1751 (= PD167) T12/4 = PD531 Phyll (= PD530), Phyl2 (= PD536) NCPPB2328 (= PD185)
Jasminum sp. Jasminum sp. Phyllirea sp. Ligustrum japonicum Thbg
Greece France (L. Gardan) France (L. Gardan) Italy
Persea americana Mill. Syringa vulgaris L.
Israel
P. syringae pv. syringae NCPPB191 (= PD321) NCPPB281 (= PD184)
europea europea europea europea europea europea europea
L. L. L. L. L. L. L.
P. syringae pv. morsprunorum NCPPB560 (= PD213)
Greece Maroc Italy (M. Scortichini) Yugoslavia Yugoslavia Italy France (L. Gardan)
U.K.
U.K.
P. syringae pv. maculicola PD236
Brassica oleracea L.
Netherlands
Pseudomonas species PD117 NCPPB1465 (= PD123)
Fraxinus excelsior L. Fraxinus excelsior L.
Netherlands
1 2
3
4
U.K.
PD, culture collection Plant Protection Service, Wageningen, the Netherlands. PDDCC, Plant Disease Division Culture Collection, Auckland, New Zealand. NCPPB, National C;;llection of Plant Pathogenic Bacteria, Harpenden, U.K. CFBP, Collection Fran~aise de Bacteries Phytopathogenes, Angers, France.
Table 1. Origin and identity of strains used in this study
Pathovars of Pseudomonas syringae and N . oleander were grown in a computer controlled glasshouse at 28°C, 85-90% RH and 10 000 Lux light after inoculation. Fatty acid analysis. Bacteria were grown for 48 h at 28°C on Trypticase Soy Broth Agar (TSBA), containing (w/v) 3% Trypticase Soy Broth (BBL) and 1.5% Bacto Agar (Difco). Circa 40 mg (wet weight) ccells were harvested from the most dilute quadrant showing confluent growth (late log phase). Whole cell fatty acids were liberated, methylated and extracted, following the method of Miller and Berger (1985). All strains were tested in duplicate. The Midi Microbial Identification System MIS (Microbial ID, Inc. Newark, DE, USA) was used. The MIS consists of a Hewlett Packard HP5890A gas chromatograph with a 25 m x 0.2 mm 5% methylphenyl silicone fused silica capillary column, H2 as carrier gas and a flame-ionization detector, an automatic sampler, an integrator and a computer. The latter identifies the fatty acids, using data of a fatty acid library and a calibration mix of known fatty acids (Microbial ID, Inc.). Statistical analysis (standard error of differences of the means) of values for some fatty acids or their ratios was performed using the program GENSTAT on a MicroVAX 3600 computer. Moreover library generating software (LGS) and a statistical program CLUS developed by Microbial ID Inc., was used for principal component and cluster analysis of strains and also for creating a reference library for P. syringae subsp. savastanoi.
Results
Physiological and biochemical tests All strains (PD 117, 123, 184, 213, 236, 321 not included, for their deviatiQns from the pattern described below see janse, 1982 a) behaved similarly in physiological and biochemical tests. They were Gram-negative, motile, non-spore forming rods, showing slow growing 'gray-white, smooth, glistening raised and circular or slightly irregular to undulate colonies on NA. Usually levan-negative on nutrient-sucrose (5% w/v) agar (PD ' 120,161,179,180,206,555 positive). Strains produced a ' weak, blue-green fluorescent, diffusable pigment on King's medium B (CFBP 1751, PD 179 and PD 534 produced a brown diffusable pigment). Glucose metabolism respiratory, oxidase negative, catalase negative. Acid was produced from D-( + )-galactose, glucose, D-( +}-ribose, sucrose (slow), D-( + }-xylose and mannitol. No acid was produced from maltose, D-( + }-raffinose (PD 161 positive), erythritol or salicin. Alkali was produced from L-( + }-tartrate (PD 120, 392, T 12/4 and T 38/1 negative). Aesculin and casein were not or only weakly hydrolyzed, arginine, gelatin, starch and arbutin were not hydrolyzed. Nitrates were not reduced, nor was H2S produced from cysteine. No growth at 37 DC or with 5% NaCl. Sodiumpolypectate was usually hydrolized at pH 5.5 (NCPPB 639, 640,1464, 2327,2328, 3278 and PD 1056 negative). Hypersensitivity was produced in tobacco (cv. Samson) leaves (NCPPB 640, PD 1265, 1266, 1296, 1300 negative, L 7/12, L 24/5, NCPPB 1481, T 12/4 doubtful). On the basis of the above mentioned tests all 52 strains were identical to P. syringae subsp. savastanoi.
Pathogenicity tests Results of pathogenicity tests on four different hosts are presented in Table 2. At least three distinct groups could 6 System. AppJ. Microbial. Vol. 14/1
81
Table 2. Results of pathogenicity tests with 52 sttains of P. syringae subsp. savastanoi on 4 different hosts Strains from F. excelsior O.europea N. 0leander3 Jasminum Ligustrum Phyllirea
F. excelsior
+
Et>1 Et>2 Et> Et>
O. europea
Pathogenicity on: N. F. interoleander media
+
Et> Et> Et>4 Et> Et>
Et>1+
+, necrotic swellings; Et>, parenchymatous galls; -, no pathoreaction Strain PD1266 and 1298 doubtful; PD1296 negative. Strain PD1300 negative. Strain NCPPB640 non pathogenic. Strain T12/4 doubtful.
~enic 2
3 4
be observed, viz. 1} strains from Oleaceae other than F. excelsior, causing parenchymatous galls on O. europea and F. excelsior (the strains from jasminum and Ligustrum, also causing parenchymatous galls on F. intermedia included), 2} strains from N. oleander causing parenchymatous galls or wartlike excrescences on N. oleander, O. europea and F. excelsior and 3} strains from F. excelsior causing necrotic excrescences on F. excelsior and O . europea. On the basis of these results strains from O. europea, jasminum, Ligustrum japonicum and Phillyrea could be classified as pv. oleae, all strains from F. excelsior as pv. fraxini and all strains from N. oleander as pv. nerii, confirming earlier results.
Fatty acid analysis Table 3 shows the mean percentages of fatty acids found in the tested strains of P. syringae subsp. savastanoi. Over 83 % consisted of the straight chain saturated and monounsaturated fatty acids 12:0 (c. 4.5%), 16:0 (c. 27%), 16:1 cis 9 (c. 31%) 18:1 cis 11 (c. 20%) and 18:0 (c. 1.5%). About 10.5% were the hydroxy fatty acids 10:0 30H, 12: 0 20H and 12: 0 30H. The cyclic fatty acid 17: 0 was found to be present in rather large amounts (c. 4%) . The above mentioned fatty acids were found in all strains. The other 14 fatty acids (in total c. 1.5%) occurred infrequently. The fatty acids 15: 0 (1 X NCPPB 2327}, 17: 0 and 19: 1 trans 7 have never been reported before to occur in P. syringae subsp. savastanoi. The fatty acids 11 : 0 iso 30H, 19: 0 iso, 20: 1 trans 11 and 20: 0 occurred in strains from N. oleander only, they were not present in all Nerium strains, however. The fatty acid pattern of the other Pseudomonas strains differed to a large extent from that of P. syringae subsp. savastanoi (Table 4) resulting in low similartiy values (below 0.5) with the P. s. subsp. savastanoi library. Statistical significant differences between concentrations of fatty acids or their ratios (Table 5) were correlated with the segregation of strains in three groups (pathovars) which was found in the pathogenicity tests.
82
J.D.Janse
Fatty acid saturated 12:0' 14:0' 15:0' 16:0' 17:0' 18:0' 20':0' branched 17:0' iso 19:0' iso cyclo 17:0' cy 19:O'cy Cll-12
count
mean%
s.d.
10'4 63
4.5 0'.2 t 27.1 t 1.5
0'04 0'.2
1
10'4 17 10'3 2 44 1
10'3 8
0'.3
3.3 0'.8
0'04
Table 3. Total cellular fatty acid profile of 52 strains* of Pseudomonas syringae subsp. savastanoi grown for 48 h on TSBA medium
s.d.
count
hydroxy 10':0' 30H 11:0' iso 30H 12:0' 20H 12:0' 30H
10'4 8 10'4 10'4
3.1 t 3.2 4.5
0'.2 0'.3
unsaturated 16:1 cis 9 18: 1 cis 11 19:1 trans 7 20':1 trans 11
10'4 10'4 5 7
31.6 19.8
3.3 5.8
56
0'.4
0'04
unknown! 19:0' cydo C 9-10' 3.7
mean%
fatty acid
0'.3
2.0'5
.t
- = not detected; t = trace
* All strains tested in duplicate.
Table 4. Percentage of several fatty acids of some Pseudomonas bacteria* related to P. syringae subsp. savastanoi Fatty acid
Bacterium
P. syringae pv. syringae (NCPPB191) P. syringae pv. syringae (NCPPB281) P. syringae pv. morsprunorum (NCPPB56D) P. syringae pv. maculicola (PD236) Pseudomonas sp. (PD117) Pseudomonas sp. (NCPPB1465) P. syringae subsp. savastanoi
12:0' 20H
16:0'
16:1 cis 9
18:0'
18:1 cis 11
3.7 3.7 3.5 3.0' 0' 0' 3.2
20'.7 25.0' 2604 25.5 25.7 19.6 27.1
36.1 34.7 30'.6 36.9 26.0' 4.9 31.6
0'.5 1.2 1.5
17.0' 17.6 17.4 18.4 18.7 32.7 19.8
1.2
0'.2 0'.9 1.5
.. Result of a test in duplicate.
Fatty acid
pv. fraxini mean %
pv. nerii mean %
pv. oleae mean %
s.e.d.
16:0' 16:1 18:0' 18:1 16:1 16:1
31.0'3 34.81 0'.98 12.69 2.82 4.24
23.85 28.56 2.31 25.86 1.16 1.88
25.47 30'.28 1.64 22.15 1.37 2.21
0'0471 0'.634 0'.162 0'.755 0'.0'89 0'.118
1
cis 9 cis 11 cis 9/18:1 cis 9+16:0'118:1+12:0' 20H
Table 5. Fatty acids or their ratios that differentiate pathovars of P. syringae subsp. savastanoi (P = 0'.0'5)1
Based on data of 50' strains tested in duplicate (deviating strains NCPPB 10'0'6 and CFBP 1838 excluded).
There was greater difference between strains from F. excelsior and those from other hosts than between strains from N. oleander and those from Oleaceae other than F. excelsior. Only four deviating strains were found, two strains from F. excelsior (NCPPB 1006 and CFBP 1848) repeatedly showed i fatty acid pattern of pv. oleae, one strain from N. oleander (PD 1300) showed a pattern of pv.
oleae and one strain from olive (NCPPB 1481) showed a pattern of pv. nerii. Principal component analysis showed that pv. fraxini formed a separate group, but that pv. oleae and nerii had an overlap for these features (Fig. 1). This suggests a closer relationship between pv. oleae and pv. nerii than between pv. fraxini and the other two pathovars.
Pathovars of Pseudomonas syringae
....c: -11.2 N
OJ
c: o
a. E
8 -13.1 ~
a.
o
I.J
c:
'.£ -15.0 -16.9
-16.6
-20.8
-24.6
•••
• • ••• •• • •
...
~----~--~~--~~--~----~
-21.3
-14.6
-7.9
-1.3
-5.4
-12.1
Principle component 1 E~clidian
distance
Fig. 1. Two-dimensional plot of principle component analysis of 52 strains of Pseudomonas syringae subsp. savastanoi (tested in duplo) showing subdivision in three fatty acid groups. These groups are corresponding to pv. fraxini (e), pv. oleae (0) and pv. . nerii (+).
Discussion This study shows once more that fatty acid analysis can be a supportive tool for the classification of plant pathogenic bacteria below species level, as was also reported by De Boer and Sasser (1986) for subspecies of Erwinia carotovora. Strains of P. syringae subsp. savastanoi allocated to pv. oleae or pv. nerii based on pathogenicity and host range could be separated into two similar groups by fatty acid analysis, confirming results of Varvaro and Sasser (1987). Wells et aL (1990) could not separate pv. oleae from pv. nerii, possibly because they used a different medium, six-day old cells, a different column and a different chromatograph. The differences in fatty acid profile of pv. oleae and pv. nerii correlated well with those found by Varvaro and Sasser (1986), even though they used 24 h old cells instead of 48 h old cells. I found, however, that many strains do not produce reproducable amounts of usable cell material after 24 h, yielding variable profiles. The acids 11: 0 iso 30H, 20: 0 and 20: 1 trans 11 were also found by me to be exclusively present in {?v. nerii. However 17: 0 iso was also found to occur after 48 h cultivation in pv. oleae and to a minor extent in pv. fraxini. While strains from L.
83
;aponicum and Jasminum showed some differences in host range as compared to strains from Olea europea, these strains all produced parenchymatous galls and showed a fatty acid profile similar to that of olive strains. Varvaro and Sasser (1987) found a similar pattern studying a larger group of strains from L. ;aponicum. Therefore no further subdivision into pathovars was thought to be useful and justified at the moment. The occurrence of new pathovars may not be excluded however since a new host has cecently been described, viz. Myrtus sp.( Gardan and Abu Ghprrah, 1987). The strains allocated to pv. fraxini on the basis of pathogenicity and host range could be clearly separated from those belonging to pv. nerii or pv. oleae, confirming results of Wells et aL (1990). Several deviating strains were found, not fitting in the pathovar pattern, as far as their fatty acids were concerned. Two strains from F. excelsior, one from France and one from the U.K. showed a pv. oleae profile. Other F. excelsior strains from these countries showed the pv. fraxini pattern. The basis and frequence of this deviation are unknown and need further study, other characters were consistent for pv. fraxini. Strains with the pathogenic phenotype of pv. nerii have been isolated from olive (Pyrolowakis and Welzien, 1974; Wilson and Magie, 1963). The strain isolated from pv. nerii phenotype was from this area, however, it did not show the pathogenic phenotype of pv. nerii. This means that deviating strains (i. e. host not correlating with pathovar classification in fatty analysis), can only be definitely identified by a pathogenicity test on different hosts. This study on fatty acids thus confirms that the existence of the three pathovars previously described Uanse, 1982 b) can be confirmed by fatty acid analysis. It must be remarked that pv. nerii and pv. oleae upon principal component analysis showed a close relationship and a partly overlap. Pv. fraxini was much more different. This correlates with the pathogenic properties and geographic distribution of the different pathovars. Pvs nerii and oleae both produce parenchymatous galls and growth promoting hormones (Surico et aL, 1985). They occur mainly in the Middle East/Southern Europe area. Pv. fraxini produces necrotic excrescences and no hormones, it occurs mainly in the MiddlelW'estern Europe area. This study also shows that the pathovars of P. syringae subsp. fraxini show more differential characters and are therefore more natural (polythetic) than originally suspected. The definition of pathovar used here is that of the Bacteriological Code (Lapage et aI., 1975): pathogenic to one or more hosts, and not the one used by Dye et al., (1980) where it has almost the definition of subspecies. When using the former definition, there would be no need for designating type strains of pathovars (pathotypes, as defined by Dye et al. 1980) since the concept of pathovars does not fall under the rules of the Code. Pathotypes, however, were mentioned by me in an abstract Uanse, 1987) and because the pathovars could be eventually elevated in rank, they are repeated here for valid publication: strain PD 118 = NCPPB 639 = ATCC 13522 as pathotype for pv. oleae
84
J.D.Janse
(also type strain for subspecies savastanoi) strain PD 181 = NCPPB 3278 as lectopathotype for pv. nerii strain PD 116 = PDDCC 7711 as pathotype for pv. fraxini When several (phenotypically or serologically) related Pseudomonas strains were compared in fatty acid analysis, P. syringae subsp. savastanoi could be clearly differentiated from them. This was also true for a saprophytic strain (NCPPB 1465) which was originally described as a deviating pathogenic strain, namely P. savastanoi subsp. fraxini (Sutic and Dowson, 1963). Moreover a hypersensitivity positive P. syringae strain isolated from a lesion on F. excelsior with a close though deviating phenotypic profile, serologically cross-reacting, but not pathogenic for ash could be easily differentiated by fatty acid analysis from subsp. savastanoi Uanse, unpubl.). These findings suggest caution when postulating all kind of transitions in the P. syringae group especially when studying strains from the phyllosphere and not from active lesions or galls (Ercolani, 1983). Acknowledgements: I wish to extend my thanks to J. H. J. Derks, B. E. Spit and W. H. van der Tuin for skilful technical assistance and Dr. j. Dekker and Dr. M. Sasser for critical reading of the manuscript.
References Comai, L., Surico, G., Kosuge, T.: Relation of plasmid DNA to indoleacetic acid production in different strains of Pseudomonas syringae pv. savastanoi. J. Gen. Microbiol. 128, 2157-2163 (1982) De Boer, S. H., Sasser, M.: Differentiation of Erwinia carotovora spp. carotovora and E. carotovora spp. atroseptica on the basis of cellular fatty acid composition. Can. J. Microbiol. 32, 796-800 (1986) , Dye, D. W., Bradbury, j. F., Goto, M., Hayward, A. c., Lelliot, R. A., Schroth, M. N.: International standards for naming pathovars of phytopathogenic bacteria and a list of pathovars names and pathotype strains. Rev. Plant Path. 59, 153-168 (1980) Ercolani, G. L.: Variability among isolates of Pseudomonas syringae pv. savastanoi from the phyllosphere of the olive. J. Gen. Microbiol. 129, 901-916 (1983) Evidente, A., Surico, G., Iacobellis, N. S., Randazzo, G.: 1'methyl zeatin, an additional cytokinin from Pseudomonas syringae pv. savastanoi. Phytochemistry 25, 525-526 (1986) Gardan, L., Abu Ghorra, M.: Identification and pathogenicity of Pseudomonas syringae pv. savastanoi. Proceedings 3rd International Working Group on Pseudomonas syringae pathovars.
Ir.
J.
Ministro da Agricultura, Instituto Nacional de Investigacao Cientifica, LisbonIPortugal (1987) janse, J. D.: The bacterial disease of ash (Fraxinus excelsior), caused by Pseudomonas syringae subsp. savastanoi pv. fraxini. I. History, occurrence and symptoms. Europ. J. Forest Path. 11,306-315 (1981a) janse, J. D.: The bacterial disease of ash (Fraxinus excelsior), caused by Pseudomonas syringae subsp. savastanoi pv. fraxini. II. Etiology and taxonomic considerations. Europ. J. Forest Path. 11,425-438 (1981 b). , janse, j. D.: The bacterial disease of ash (Fraxinus excelsior), caused by Pseudomonas syringae subsp. savastanoi pv. fraxini. III. Pathogenesis. Europ. J. Forest Path. 12, 218-231 (1982 a) janse, j. D.: Pseudomonas syringae subsp. savastanoi (ex Smith) subsp. nov., nom. rev., the bacterium causing excrescences on Oleaceae and Nerium oleander L. Int. J. System. Bact. 32, 166-169 (1982 b) janse, j. D.: Lesser known infections of several forms of Pseudomonas syringae. Proceedings 3rd International Working Group on Pseudomonas syringae pathovars. Ministra de Agricultura, Instituto Nacional de Investigacao Cientifica, LisbonIPortugal (1987) Lapage, S. P., Sneath, P. H. A., Lessel, E. F., Skerman, V. B. D., Seeliger, H. P. R., Clark, W. A.: International code of nomenclature of bacteria. Bacteriological Code (1975 revision), Washington DC., American Society of Microbiology 1975 Miller, L., Berger, T.: Bacterial identification by gas chromatography of whole cell fatty acids. Hewlett-Packard Application Note 228-41, 8 pp. (1985) Pyrowolakis, E., Welzien, H. c.: Studies on the distribution of olive-knot, induced by Pseudomonas savastanoi (Sm.) Stev. in the Greek island of Crete. Phytopath. Mediterr. 13, 118-120 (1974) Smidt, M., Kosuge, T.: The role of indole-3-acetic acid accumulation by alpha methyl tryptophan-resistant mutants of Pseudomonas savastanoi in gall formation on oleanders. Physiol. Plant Path. 13,203-214 (1978) Surico, G., Iacobellis, N. S., Sisto, A.: Studies on the role of indole-3-acetic acid and cytokinins in the formation of knots on olive and oleander plants by Pseudomonas syringae pv. savastanoi. Physiol. Plant Path. 26, 309-320 (1985) Sutic, D., Dowson, W. J.: Microbiological characteristics of some isolates and varieties of Pseudomonas savastanoi (Smith) Stevens. Phytopath. Zschr. 49, 156-160 (1963) Varvaro, L., Sasser, M.: Fatty acid profiles of Pseudomonas syringae pv. savastanoi. Proceedings 3rd International Working Group on Pseudomonas syringae pathovars. Ministro de Agricultura, Instituto Nacional de Investigacao Cientifica, LisbonIPortugal (1987) Wells, J. M., Casano, F. J., Surico, G.: Fatty Acid composition of Pseudomonas syringae pv. savastanoi. J. Phytopath. (1990), in press Wilson, E. E., Magie, A. R.: Physiological, serological and pathological evidence that Pseudomonas tonelliana is identical with Pseudomonas savastanoi. Phytopathology 53, 653-659 (1963)
D. janse, Department of Bacteriology, Plant Protection Service, P.O. Box 9201, 6700 HC Wageningen, the Netherlands